Multi-Class Eye Disease Classification Using Deep Learning

Abstract

Retinal fundus diseases affect a significant portion of the population and can lead to vision loss or blindness if left untreated. In this work, we developed a deep learning-based system for the multi-class classification of eye diseases using a large dataset of retinal images. The system can accurately detect and classify normal, diabetic retinopathy, glaucoma, and cataracts with an overall accuracy rate of 92%. Additionally, it provides users with detailed information on the symptoms and descriptions of the classified diseases and allows for easy image uploading and classification tracking. Our work aims to address the need for an accurate and accessible tool for identifying eye diseases, enabling earlier intervention, and improving patient outcomes. We also aim to advance the field of ophthalmology by demonstrating the potential of deep learning algorithms to enhance disease diagnosis and management